Zhongjie Yu

Title: Zhongjie Yu and Edge-Device Machine Learning

Introduction

Zhongjie Yu is an inventor based in Bellevue, Washington. His work includes an invention for developing and deploying machine-learning models across edge devices, with an emphasis on reducing the need for manual labeling and enabling models to improve without human intervention.

Latest Patents

Yu is associated with 1 patent, titled “Self-supervised collaborative approach to machine learning by models deployed on edge devices.” The invention describes a surveillance system in which edge devices collaborate with a server system through self-supervised learning and knowledge distillation. This approach is designed to train and update deployed models as needed, improving inference quality while reducing or eliminating manual labeling.

Career Highlights

Yu works at Wyze Labs, Inc. His patent reflects work on practical machine-learning methods for systems that distribute model training and updates between edge devices and server infrastructure.

Collaborations

The provided information identifies Lin Chen and Mohammadmahdi Kamani as Yu’s coworkers. The patent’s collaborative approach also involves edge devices and a server system working together to support model training and updates.

Conclusion

Zhongjie Yu’s patent focuses on making machine-learning models for edge devices easier to train and improve through self-supervised learning and collaboration with server systems.